The studies proposed seek to elucidate rules controlling the number and qualitative composition of photoreceptor synapses. A major objective is to understand the control of the assembly of multiple-contact synapses (dyads, triads, etc.) from a detailed knowledge of their normal development sequence and through controlled alterations either to the number, size or combination of their cellular participants. For this, the photoreceptor synapses in cartridges of the first optic neuropile of Musca and Drosophila have been chosen as a model, since each cartridge incorporates a small, fixed set of identified neurons. Specific objectives are: (a) to undertake an analysis of the complete connectivity matrix of all elements of the cartridge from serial EM. Of particular interest for Musca are the number of forward vs feedback synapses between reciprocally-connected elements, needed to consider a particular neuron in both pre- and postsynaptic roles during synaptogenesis. The analysis in drosophila is without serious precedent and will serve as baseline information from which to analyze visual mutants with connectivity defects. (b) to examine the combinatorial preferences of neurons during synaptogenesis, following two types of perturbation. Laser-induced deletions of the developing or adult lamina will procure the loss of one or more postsynaptic cells, to test the opportunity for additions, deletions, or substitutions of postsynaptic partners at photoreceptor tetrad synapses. The effects of similar deletions will also be examined in cartridges of the Drosophila mutant, Vam, where one or both of a pair of postsynaptic neurons in the adult undergo(es) degeneration with a predictable time-course. (c) to examine quantitative aspects of synaptogenesis in two studies: on the effects of visual experience on the frequency of photoreceptor synapses in Musca, to be studied using single-section quantitative EM analysis. Given that the synapse signals contrast, the effects of flickering lights will be examined. Secondly, synapse spacing will be analyzed at different developmental ages, from serial EM and computer reconstruction of normal photoreceptor terminals and postsynaptic dendrites, to examine the relationship between synaptogenesis and dendritic morphogenesis. The studies proposed here because they aim at producing a basic model of synaptogenesis applicable to multiple-contact synapses, such as dyads and triads, should contribute to a general understanding of the perturbations in disease states to which synapses are susceptible during their development.